High speed downlink packet access (HSDPA) is a feature that was introduced in Release 5 of the third generation partnership project (3GPP) specification. HSDPA achieves maximum spectral efficiency using three key concepts: adaptive modulation and coding (AMC), fast physical layer retransmissions, (i.e., hybrid automatic repeat request (HARQ)), and fast Node B scheduling.
A handover is a process in which a wireless transmit/receive unit (WTRU) switches from one cell to another without service interruption. FIG. 1 shows a conventional handover from one cell to another. In HSDPA, a WTRU 102 monitors a channel in a single cell, which is called a “serving high speed downlink shared channel (HS-DSCH) cell.” When a handover occurs, the WTRU 102 needs to switch to a new serving HS-DSCH cell (target cell 106) and stop communication with an old serving HS-DSCH cell (source cell 104). This procedure is also called serving HS-DSCH cell change.
A WTRU continuously measures the signal strength of neighboring cells. Once the signal strength measured on the monitored common pilot channel (CPICH) of the neighboring cell exceeds that of the serving cell, (i.e., event 1D), the WTRU indicates a radio network controller (RNC) of the change of best cell. The change of best cell is reported from the WTRU to the RNC via a radio resource control (RRC) measurement report. The measurement report contains a measured value and a cell identity (ID). The RNC then makes a decision whether a serving HS-DSCH cell change should take place.
In order to initiate the serving HS-DSCH cell change, a serving radio network controller (SRNC) requests a controlling radio network controller (CRNC) to allocate HS-DSCH resources, (such as HS-DSCH radio network temporary identity (H-RNTI), high speed shared control channel (HS-SCCH) codes, HARQ resources, etc.), for the WTRU in the target cell via radio network subsystem application part (RNSAP) and Node B application part (NBAP) messages. Once the HS-DSCH resources are reserved, the CRNC provides all the information to the SRNC which in turn sends an RRC handover message to the WTRU. The RRC message that may be used to indicate a serving HS-DSCH cell change to the WTRU includes, but is not limited to, a physical channel reconfiguration message, a transport channel reconfiguration message, a radio bearer reconfiguration message, and an active set update message.
The RRC handover message provides the WTRU with radio access parameters required for the WTRU to start monitoring the target cell. In addition, the RRC message may provide an activation time at which the handover should take place.
Two types of handovers exist: synchronized and unsynchronized handovers. In an unsynchronized handover, the network and the WTRU do not activate and switch the resources at the same time. The activation time for the WTRU given in the handover command is set to “now.” This reduces the delays associated with the handover procedure. However, it increases the probability of data loss.
In a synchronized handover, the network and the WTRU activate and switch the resources simultaneously. The network has to set the activation time to a conservative value to account for any kind of delays such as scheduling delay, retransmissions, configuration time etc. Even though the synchronized handover minimizes data losses it may result in a longer delay.
Conventionally, the RRC handover message is sent to the WTRU via the source Node B. The delay associated with the serving HS-DSCH cell change procedure may cause the handover message to fail, resulting in an unacceptable rate of dropped calls. Several proposals have been introduced to optimize the serving HS-DSCH cell change procedure.
In accordance with the proposals, a WTRU and a Node B may be preloaded (pre-configured) with HS-DSCH related configuration. When a cell is added to an active set and if the RNC decides the cell can be added to the “HSDPA active set,” a WTRU and a Node B are pre-configured with the radio link reconfiguration prepare/ready phase. When a change in best cell occurs the target Node B may be commanded to start scheduling the WTRU with the radio link reconfiguration commit/start phase. This enables the WTRU and the Node B to start communicating faster.
A WTRU may monitor source and target cell HS-SCCHs in parallel. Upon change of best cell, the WTRU sends a measurement report 1D message. After waiting for a configurable amount of time the WTRU starts monitoring the pre-loaded target cell HS-SCCH in addition to the HS-SCCH of the source cell. With this scheme, service discontinuity may be reduced.
A target Node B may be implicitly re-pointed at first scheduling occurrence. When an RNC authorizes the handover and the target Node B is configured and ready, the target Node B may schedule the WTRU on one of HS-SCCHs which are monitored by the WTRU. The first scheduling occurrence from the target Node B implicitly confirms the successful handover. To avoid packet loss the source Node B may provide the RNC a status of how much data still needs to be transmitted.
The handover (or re-pointing) indication may be sent over the target Node B via an HS-SCCH order, via a new physical layer channel, or via a serving cell change channel (SCCCH) that uses the same channelization code as an enhanced dedicated channel (E-DCH) relative grant channel (E-RGCH) and an E-DCH HARQ indicator channel (E-HICH) but with a different signature sequence. The WTRU acknowledges the handover indication by changing the uplink scrambling code or by using a special value of a channel quality indicator (CQI), (e.g., 31).
In accordance with another proposal, following the above proposals, a WTRU requirement may be limited to monitor only one HS-SCCH of the pre-allocated/non serving cell that triggered event 1D. Event 1A and 1B may be reused with different parameter values to create an “HS-DSCH serving candidate set”, which is a subset of cells that are almost as good as the best cell. If a cell within an active set becomes almost as good as the active set, event 1A* is triggered. The target Node B configurations are pre-configured and the HS-SCCH codes are allocated. The first HS-SCCH code in the list is called a primary HS-SCCH code. The pre-configuration is sent to a WTRU. When an event 1D occurs, the WTRU only starts monitoring the primary HS-SCCH of the target Node B in addition to the source Node B's HS-SCCH. Upon reception of first scheduling of the target Node B, the WTRU stops receiving HS-DSCH from the source cell. The target Node B considers the reception of a positive acknowledgement (ACK) from the WTRU as an indication of successful handover.
In accordance with yet another proposal, a handover command, (i.e., the handover message), may be sent over the target cell using a common channel with known configuration. WTRUs may use a common HS-DSCH radio network temporary identity (H-RNTI) to monitor HS-SCCH on the target cell. The common information can be broadcasted over system information blocks (SIBs) or configured via dedicated RRC messages. To increase the reliability of the handover message, the network may send the message over both source and target cells.
In order to allow a WTRU in a CELL_DCH state to receive a serving HS-DSCH cell change message over the target cell using common resources, the WTRU has to be able to read the SIB to acquire the HS-DSCH system information. In accordance with the conventional 3GPP specification, WTRUs are not allowed to read the required SIBs while in CELL_DCH. In addition, since the broadcasted message over the SIBs is only repeated according to a repetition factor, a WTRU may not be given enough time to acquire the information. This may result in a failure to receive the handover command. In addition, when dedicated pre-loaded resources are used to enhance the serving HS-DSCH cell change, the change of the best cell outside of the active set may not be performed using such enhancements.